Meta told Nebraska officials that a vulnerability in an AI-assisted Instagram account recovery system allowed unauthorized third parties to send password reset links to email addresses not associated with targeted accounts. Reporting from 404 Media, TechCrunch and BleepingComputer says attackers used the flaw to tak...
Meta notified the Nebraska Attorney General that attackers exploited a vulnerability in an Instagram AI-assisted account recovery system to reset user accounts.
In a data breach notification filed with the Nebraska Attorney General, Meta said a vulnerability affected Instagram’s AI-assisted High Touch Support account recovery system. According to the notice, unauthorized third parties were able to perform password resets by sending reset links to email addresses that were not associated with the targeted Instagram accounts.
Meta’s filing describes the incident as a flaw in account recovery functionality, not as a breach of the company’s underlying infrastructure. But the practical effect described in the notice was serious: reset links could be routed to email addresses that did not belong to the affected accounts.
BleepingComputer reported, citing Meta’s disclosure, that 20,225 Instagram users were affected. The outlet also reported that the affected accounts did not have two-factor authentication enabled, and that users with two-factor authentication were not affected by this incident.
404 Media reported that hackers used Meta’s AI support chatbot to access high-profile Instagram accounts by asking it to change the email address tied to a target account. TechCrunch similarly reported that attackers told Meta’s AI chatbot they owned targeted accounts, linked attacker-controlled email addresses, and then reset passwords to take control.
Those accounts align with Meta’s notice to Nebraska officials, which says the vulnerability allowed password reset links to be sent to email addresses that were not associated with the affected Instagram accounts.
Meta’s disclosure does not frame the issue as a chatbot merely generating unsafe text. The more consequential problem was that an AI-assisted support system was connected to account recovery actions. In that setting, the system’s ability to update or validate account contact information becomes security-sensitive.
The public record points to a failure around account recovery authorization. If a recovery flow allows a reset link to be sent to an email address that is not already linked to the account, that flow can become a path to account takeover.
That distinction matters for companies deploying AI in customer support. A chatbot that only answers questions creates different risks from an AI-assisted system that can call internal tools, alter account records or trigger password resets. When AI systems can take operational actions, their permissions and checks need to be designed like those of other privileged software clients.
TechCrunch reported that Instagram was alerting users who had been targeted. Meta’s Nebraska notice indicates the issue was tied to accounts without two-factor authentication, reinforcing the importance of stronger login protection for consumer accounts.
The available filings and reports do not show that attackers needed to defeat a complex model safety system. The documented issue was that an AI-assisted recovery mechanism could be used to send reset links to unassociated email addresses.
For companies building AI-assisted support, the lesson is direct: account recovery should not rely only on conversational claims of ownership. Systems that can change contact details, initiate password resets or otherwise act on accounts need independent authorization checks, scoped permissions and audit trails.
Support automation may reduce workload, but account recovery remains one of the highest-risk areas for consumer platforms. Meta’s disclosure shows that when AI-assisted systems are allowed to perform recovery actions, authorization design is as important as the model interface itself.
Meta notified the Nebraska Attorney General that attackers exploited a vulnerability in an Instagram AI assisted account recovery system to reset user accounts.
Meta disclosed an AI assisted recovery flaw In a data breach notification filed with the Nebraska Attorney General, Meta said a vulnerability affected Instagram’s AI assisted High Touch Support account recovery system.
According to the notice, unauthorized third parties were able to perform password resets by sending reset links to email addresses that were not associated with the targeted Instagram accounts.
Continue reading
